sergiotarxz/nuttx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

SMP: Fix an error in critical section logic when performing a context switch from an interrupt handler. The g_cpu_irqset bit was not being set for the CPU so other CPUs did not know about the critical section.

Browse Source
This commit is contained in:
Gregory Nutt 2017-01-13 06:48:10 -06:00
parent 4ede950039
commit 9ce4022096
4 changed files with 83 additions and 32 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
sched/irq/irq.h
View File

@ -1,7 +1,7 @@
/****************************************************************************
* sched/irq/irq.h
*
* Copyright (C) 2007, 2008, 2013-2014 Gregory Nutt. All rights reserved.
* Copyright (C) 2007, 2008, 2013-2014, 2017 Gregory Nutt. All rights reserved.
* Author: Gregory Nutt <gnutt@nuttx.org>
*
* Redistribution and use in source and binary forms, with or without
@ -72,6 +72,10 @@ extern volatile spinlock_t g_cpu_irqlock SP_SECTION;
extern volatile spinlock_t g_cpu_irqsetlock SP_SECTION;
extern volatile cpu_set_t g_cpu_irqset SP_SECTION;
/* Handles nested calls to enter_critical section from interrupt handlers */
extern volatile uint8_t g_cpu_nestcount[CONFIG_SMP_NCPUS];
#endif
/****************************************************************************

31
sched/irq/irq_csection.c
View File

@ -1,7 +1,7 @@
/****************************************************************************
* sched/irq/irq_csection.c
*
* Copyright (C) 2016 Gregory Nutt. All rights reserved.
* Copyright (C) 2016-2017 Gregory Nutt. All rights reserved.
* Author: Gregory Nutt <gnutt@nuttx.org>
*
* Redistribution and use in source and binary forms, with or without
@ -66,16 +66,10 @@ volatile spinlock_t g_cpu_irqlock SP_SECTION = SP_UNLOCKED;
volatile spinlock_t g_cpu_irqsetlock SP_SECTION;
volatile cpu_set_t g_cpu_irqset SP_SECTION;
#endif
/****************************************************************************
* Private Data
****************************************************************************/
#ifdef CONFIG_SMP
/* Handles nested calls to enter_critical section from interrupt handlers */
static uint8_t g_cpu_nestcount[CONFIG_SMP_NCPUS];
volatile uint8_t g_cpu_nestcount[CONFIG_SMP_NCPUS];
#endif
/****************************************************************************
@ -303,6 +297,13 @@ try_again:
/* In any event, the nesting count is now one */
g_cpu_nestcount[cpu] = 1;
/* Also set the CPU bit so that other CPUs will be aware that this
* CPU holds the critical section.
*/
spin_setbit(&g_cpu_irqset, cpu, &g_cpu_irqsetlock,
&g_cpu_irqlock);
}
}
else
@ -455,18 +456,22 @@ void leave_critical_section(irqstate_t flags)
}
else
{
/* No, not nested. Release the spinlock. */
/* No, not nested. Restore the g_cpu_irqset for this CPU
* and release the spinlock (if necessary).
*/
DEBUGASSERT(spin_islocked(&g_cpu_irqlock) &&
g_cpu_nestcount[cpu] == 1);
spin_lock(&g_cpu_irqsetlock); /* Protects g_cpu_irqset */
if (g_cpu_irqset == 0)
FAR struct tcb_s *rtcb = this_task();
DEBUGASSERT(rtcb != NULL);
if (rtcb->irqcount <= 0)
{
spin_unlock(&g_cpu_irqlock);
spin_clrbit(&g_cpu_irqset, cpu, &g_cpu_irqsetlock,
&g_cpu_irqlock);
}
spin_unlock(&g_cpu_irqsetlock);
g_cpu_nestcount[cpu] = 0;
}
}

37
sched/sched/sched_addreadytorun.c
View File

@ -1,7 +1,7 @@
/****************************************************************************
* sched/sched/sched_addreadytorun.c
*
* Copyright (C) 2007-2009, 2014, 2016 Gregory Nutt. All rights reserved.
* Copyright (C) 2007-2009, 2014, 2016-2017 Gregory Nutt. All rights reserved.
* Author: Gregory Nutt <gnutt@nuttx.org>
*
* Redistribution and use in source and binary forms, with or without
@ -316,19 +316,40 @@ bool sched_addreadytorun(FAR struct tcb_s *btcb)
&g_cpu_schedlock);
}
/* Adjust global IRQ controls. If irqcount is greater than zero,
* then this task/this CPU holds the IRQ lock
/* Adjust global IRQ controls. This works differently if we are
* performing a context switch from an interrupt handler and the
* interrupt handler has established a critical section. We can
* detect this case when g_cpu_nestcount[me] > 0:
*/
if (btcb->irqcount > 0)
if (g_cpu_nestcount[me] <= 0)
{
spin_setbit(&g_cpu_irqset, cpu, &g_cpu_irqsetlock,
&g_cpu_irqlock);
/* If irqcount is greater than zero, then this task/this CPU
* holds the IRQ lock
*/
if (btcb->irqcount > 0)
{
spin_setbit(&g_cpu_irqset, cpu, &g_cpu_irqsetlock,
&g_cpu_irqlock);
}
else
{
spin_clrbit(&g_cpu_irqset, cpu, &g_cpu_irqsetlock,
&g_cpu_irqlock);
}
}
else
{
spin_clrbit(&g_cpu_irqset, cpu, &g_cpu_irqsetlock,
&g_cpu_irqlock);
/* Sanity check. g_cpu_netcount should be greater than zero
* only while we are within the critical section and within
* an interrupt handler. If we are not in an interrupt handler
* then there is a problem; perhaps some logic previously
* called enter_critical_section() with no matching call to
* leave_critical_section(), leaving the non-zero count.
*/
DEBUGASSERT(up_interrupt_context());
}
/* If the following task is not locked to this CPU, then it must

41
sched/sched/sched_removereadytorun.c
View File

@ -1,7 +1,7 @@
/****************************************************************************
* sched/sched_removereadytorun.c
*
* Copyright (C) 2007-2009, 2012, 2016 Gregory Nutt. All rights reserved.
* Copyright (C) 2007-2009, 2012, 2016-2017 Gregory Nutt. All rights reserved.
* Author: Gregory Nutt <gnutt@nuttx.org>
*
* Redistribution and use in source and binary forms, with or without
@ -260,23 +260,44 @@ bool sched_removereadytorun(FAR struct tcb_s *rtcb)
&g_cpu_schedlock);
}
/* Interrupts may be disabled after the switch. If irqcount is greater
* than zero, then this task/this CPU holds the IRQ lock
/* Adjust global IRQ controls. This works differently if we are
* performing a context switch from an interrupt handler and the
* interrupt handler has established a critical section. We can
* detect this case when g_cpu_nestcount[me] > 0:
*/
if (nxttcb->irqcount > 0)
if (g_cpu_nestcount[me] <= 0)
{
/* Yes... make sure that scheduling logic knows about this */
/* If irqcount is greater than zero, then this task/this CPU
* holds the IRQ lock
*/
spin_setbit(&g_cpu_irqset, cpu, &g_cpu_irqsetlock,
&g_cpu_irqlock);
if (nxttcb->irqcount > 0)
{
/* Yes... make sure that scheduling logic knows about this */
spin_setbit(&g_cpu_irqset, cpu, &g_cpu_irqsetlock,
&g_cpu_irqlock);
}
else
{
/* No.. we may need to release our hold on the IRQ state. */
spin_clrbit(&g_cpu_irqset, cpu, &g_cpu_irqsetlock,
&g_cpu_irqlock);
}
}
else
{
/* No.. we may need to release our hold on the irq state. */
/* Sanity check. g_cpu_netcount should be greater than zero
* only while we are within the critical section and within
* an interrupt handler. If we are not in an interrupt handler
* then there is a problem; perhaps some logic previously
* called enter_critical_section() with no matching call to
* leave_critical_section(), leaving the non-zero count.
*/
spin_clrbit(&g_cpu_irqset, cpu, &g_cpu_irqsetlock,
&g_cpu_irqlock);
DEBUGASSERT(up_interrupt_context());
}
nxttcb->task_state = TSTATE_TASK_RUNNING;